Nsl-14 g6

OUTGASSING TESTS ON IRAS SOLAR PANEL SAMPLES

G. Premat* A. Zwaal* N.H. Pennings**

I. INTRODUCTION/

i.1 .Purpose

The IRAS satellite.(InfraPed :,_tronomical Satellite) is a joint venture

between the Netherlands, United States and the United Kingdom to inves-

t/gate the infrared sources in space by means of a telescope, which is

cooled to superfluid helium temperature (about + 2 - 10K).

Due to the low operation temperature the risks of contamination,

particularly of the optical com[_0nents, are high.The efforts to diminish the contamination hazard to a minimum include

selection of low out_Tjassing materials, spr-cial treatment after manu-

facturing and speci;1 spacecraft handling procedures. All the consti-

tuting materials of the satell±te are potential sources of contamination.

Due to their location, their relatively high temperature and the many

organic materials contained, the solar panels require special considera-tion.

At the request of Fokker, prime contractor for the IRAS project and

manufacturer of the solar panels, the Materials Section of ESTEC carried

out several outgassing tests on representative solar panel samples with

a view to determining the extent of contamination that could be

expected from this source.

1.2 Purpose of tests carried out in ESTEC

The materials for the construction of the sol_r panels have beenselected as a result of contamination obtained in "Micro VCM" tests.

This method is a "screenilg test" and gives only the "qualified" or

"non-qualified" results according to criteria used by NASA as well as

by ESA (1% TMI, and 0.1% CVCM).

This test method is intended to provide an accelerated means of deter-

mining the total mass loss of material_ and their contaminationattributes in exposure to vacuum (<10-Vtorr) at a temperature of 125°C.

This method might give misleadin_ results in case of those materials,whose outgassing kinetics at 125 C are diffe_'ent from those at the

temperature of intended use.• 0 0 . • O

The comparlson of materlals prop@rtles is only valld at 125 C. TheQ • • 0 • ° • .

measurement for collectors at 25 C wlth sLm_lar stlcklng coefflclents.

It is considered, however, that the method of using the defined

screening operation is a conservative one. The risk is that a few

matcrials which have acceptable p_ operties (at the intended temperature

of use) will be eli_t'%nated because their properties are not satisfactory

0t the test temperature of 125°C. It was, however, necessary to know

the kinetics of outgassing (i.e. total mass loss of mate_'ials and

•ESTEC Noor_ijk The Netherlands

•*Fokker S_ace Division Schiphol The Netherlands

227

i

1981005626-236

https://ntrs.nasa.gov/search.jsp?R=19810005644 2020-04-10T20:04:24+00:00Z

la )_

A_ ./"/

II

ic id

Figure 1

228

s

1981005626-237

ccndensable volatile materials coming from selected products versus

time, temperature of solar panels and different temperature configura-

tions of telescope) to determine the contamination risks during themission of the satellite.

2. DESCRIPTION OF SAMPL :S

The six test samples have been manufactured like solar panels of a

flight model (Figure i).

3. DESCRIPTION OF TEST EQUIPMENT

The equipment developed at

ESTEC to perform the out- vsoc-i OUTC_SS_NGSVSTEM

gassing measurements makes

use of the advantages offered

combining QCMs (Quartz _|_bu_,

Crystal Balance) with a con- _ _ . _----_,_s_,.ventional vacuum balance. ____This system (Figure 2) con-

sists of a Sartorius 4201 ,0____ %_._-,_smv_e,mvacuum balance with magnet- s,s,,. _ Z TO_,Sat_IUQEically suspended sample

I

hangers. ! .--_--_vz _vsT_s

The balan__ has a sensitivity -- _ ! _=_D===-_,_of 1 x 10--g but due to the' -----T-'-T-_vibrations as well as temper- I u-r--coo_too,_._ _.ATEature and humidity variations

in laboratories, the__ccuracy _ _-_i_

is limited to 1 x i0 -g. The __maximum load of the balance

is 25 g and the temperature

is controlled by an oven con-

taining the samples.

The three Celesco QCMs can

measure the amount of

materials cgndensed oll the Figure 2

exposed _rystal_down to a few

time 10--9g/cm-'2(sen_itivity given by the manufacturer:8.9 x i0- g/cm" /mV-" output).

These QCMs can be cooled down from an ambient temperature to liquid

nitrogen temperature.

The condensor plate (with holes in front of QCMs) is also controlled

at the lowest QCM temperature and has a diameter of 146 mm (area + 175 cm 2'_

The contaminants trapped on the plate can be collected for analysis

(i.e. infrared spectroscopy).

The pumping system is a Varian VTI02 ion pump system, equipped with a

quadruple mass spectrometer permitting analysis cf outgassing products

" which do not condense on the condensor plate (i.e. mainl,, water,

solvents and atmosph,:ric gases). Also pressures down to at least 10 -9

tort can be obtained.

229

|

1981005626-238

4. SAMPLE PREPARATXON

Each sample has been conditioned for at least 24 hours at a standard

temperature of 20°C and a standard humidity of 65% before evacuating

the test chamber and starting the test. This is similar to the condi-

tioning used in the Micro VCM procedure (PSS-02T).

a. Clean the [X_Ms and cooled plate with a mixture of IPA and freon TF(50%-50%).

b. Mount the sample on vacuum balance and switch on.c. Close the oven.

d. Pump down.

e. Cool the QCMs and cooled _late to the intended temperature whenthe pressure is about I0- torr.

f. When the QCMs and cooled plate temperatures are correct, start

heating the oven.

In the normal sequence of operations, there was a minimum time between

the actions; ho%_ver, in some cases the time between the actions

increased due to pumping failure, cooling failure, stabilization, etc.(see Table i).

For the tests 1046/1047 and 1049 the cooling of the QCMs and heating

of the sample have been started together under dry nitrogen at

atmospheric pressures before pump down. The QCMs and the condensor

plate are maintained at temperature as described in Table 2. Through

the choice of these temperatures, a discrimination can be made between

outgassing products with respect to their vapor pressure. The composi-

tion of the lightest fractions has been analyzed using a mass

spectrometer in the pumping line.

5. MASS "PECTROMETER DATA

The analysis of gases coming out of samples has been carried out with a

quadruple spectrometer (see Table 3) This analysis has only been

carried out with the experiment 1042 (sample 3) because of failure

of the spectrometer.

5.1 Infrared anal_sis of contaminants trapped on cooled plate

After each test the composition of the organic products condensed on

the cooled plate have been analyzed by infrared spectrometry method.The results are shown in Table 4.

6. METHODS OF CALCULATION

The different parameters were collected continuously on si,_ channels

recorder and the recordings were analyzed by hand to measure the

signal V.

a. In mV for the total mass loss versus time.

b. In mV for volatile condensable _, versus time,

230

1981005626-239

The TML is calculated acoording to the following formula:

% TML = V(atm + V buo_anc_ effect - Vt) x S x i00Mo

whe re :

V atm = signal at atmospheric pressure

V = buoyancy effect

Vt = signal during test versus time

S = sensitivity lO-4g mV-1

Mo = initial weight before testing.

The CVCM is calculated according to the formula:

% CVCM = (Vt - Vo) x 8.86 x 10-9 x 175x 100

Mo

where :

Vo = signal at time o

Vt = signal during test versus time

8.86 x i0 g/cm-2/mV -1 sensitivity of QCMs (given by manufacturer)

175cm 2 = _rea of condensor plate

Mo = initial weight of sample before test in grammes.

These data are listed and stored in the computer for calculations.

buoyancy effect=buoyancy of balance + buoyancy of sample

(not including the air contained in the sample).

z_ Vc=the lowest signal observed after heating the sample.

The time zero has been taken from the heating of the sample for the

first four tests and at pump down time for the other three tests.

The zero value of the total mass loss has been measured from the

initial signal of vacuum balance as soon as the sample was mounted

corrected from the buoyancy effect measured after each test.

Due to vibrations of the vacuum balance during the mounting of the

sample and the pump down, the measurement of buoyancy effect is very

difficult and an errur of a few mg can be done.

The zero value of the QCMs has been taken at the lowest value of

signal observed after cooling of the QCMs and heating of sample,

but this can be slightly overestimated because some contamination may

already have occurred during the cooling.

231

|

1981005626-240

3a

2,ei"_- _ _ 3 I I

t,"va'3 - :s.u -_ - yI I

t.lli"_l- 3.41'-_I 41 _J 3.1_"3,

8• _ o 3E':S•

a,lkl"_ o _lollE-_,

2"11"3" ' 2.I_'_' if"St'_"

/l.II'_" 2.4E'3.

8.IIE"3- _.2t-_

I,.71"3 • 2E'3" '

,.1_," ,.OE'3" /

I..4/"3"|.I/"$" Z.6r..'_]/

,._I/',, |.4E tlI

""+"" , ! -+ I ,| .111_$•

,._, , /,, _, ii, +, ,io''_ *_'o +o ,, ,o .o _+o -J,_

3b 3c

Figure 3

232

1981005626-241

3d 3e

4PJ'3 l I " I I

-- : +_;_e'+-1111"3 -- _

• - q o31"3

31111"3

!!!i:3_3 I I o ....... I3.u'3- I I

3,.T'3-

. + e_I 81' $

a (rdr 3,_

II41"3 Oril

1"_'31 ItiP'S. -/-- e.15l- I'

1 • qll"3' f

1. Irdr'3, e.lo- l,+.+'3. I,+._'3, O.aO- I

I11"3' O.l*"

01"4' I

u-, l _ o.,',, I I+o .o _o o ,6o ,io _o io +o _o ir am +o 6o---"_.ZAlU 114,1 _Ov "in _n_.e('_ev-l} JIMI 10414 PIlL 7;a._/,,o.,_J

3f 3g

Figur,_ 3

233

1981005626-242

% %4.:ur-3q I • " "

40t"33.oi"3 j 4. ,-,'3 - I

3.61"3 1 4.4E"3. ,

4aE33.41"3 I • " "

.,-31 li 41"3._"3 -- -- ++ 31£ 3• - +;_el : I 3.,IE03-

II 3;11"I-

;liE, . -

5, 41"3 31"3'281"3

2 _r 3 2.1rdr3.- .

8/3 ;141"3

1.fiE'3" 2.2E'3'

1.61"-3' _ti-3'+..811"3'

1.41"3' I lie 3

I 21" 3 .t.41"3'I• °121"3 'I , ----I_'3'o _o to :o ,o Io 6, ;o io _o 4o _o _o 7o oo

ZIIII 1544 _ "Ill 7",,,.t./M_',J) ZRRG tO4dl C_R_lr "ltO ,,'P',ooOCNo.,,J)

3h 3i

3 41 3 e/e

o+a.- I "_3 I Io.+*- I I =dr'=. I 4 .....0..,,--- ---r :tu.:J.- . l

i to.++- I I.M'3"o.m- I +_: ;O.3"t. I " "'

o.x. I t._-I.o.w. I 8_'=..... I...... -t .......o.34. I : .m,'_. I I

o._. I t .u"a. I I

0.31. l , .,tl'3. I Io.3,, I Io.: L I_ L._ _ :.u':.

I I I I Io._. I I I ur'=...... I..... -I ......

o.ar:'- I I I K',.

J ,t. I+ "'o lo +o io 16o l :_ o +o

_'5mB¢#,w,,,'l) ;Rill 11411 CI4:R 0118 ..";,o,,,,',,,,,,..o_,,rlll_ 1141 'I'm.

3)e..+j

Figure 3

234

|

1981005626-243

41"S' %

".n _ i F-I

_.4r-a. I o.w

:,'.u"_. ! I "'1 A#O'V '31"3-

"" ""1/ :a.lu_. I I

• ._r',- I I "'lf IIa.u"_- l I o.a--11_r'_ ..... I...... I ...... I

'" ' ' ''1 :i.6r'_. I I

_._'_. l l ""I II .u-3. I Ie.IH_ur'1. I

Xil4 lees "I'II I_...{.,,*,',,I #JlileI141 Int. _,,.r(N.,.otJ

31 3m

e.m- ":. I I e.tl. % I I

o._- I o.a_. I.- -- -- 4- -- -i"

e°M- |o,I •

O.It" ---- -- -- O.ll-

l.II" O.l"

.., .. LtII.Ill" I '

I.IIIo,_t- I I

o.ss, o._l-_ I i

...... I Io it " - r r

'."- ' ' i. I. ". io' !Z/IMe 1147 _ ;'*,_vo-/,,e,*,_) JrJIMO11411 _ _'"_{Mem-r_

3n 30

Figure 3

235

1981005626-244

7. COMPUTER CALCULATION METHOD

The method =_ed is linear regression by equal incuements. It is

based on several tests of change_ of variable, lead'+ng to a linear

relation between the new variables and the necessary parameters by

least square fit.

The calculation program is written in ArL. It _onsists of a series

of sub-programs corresponding to different equations, which are

tested successively by comparison with experimental dat2_, p_ticularlythe power function Y = aT- and the exponent function y = ae ".

Both equations are given; the correlation with the power function

gives normally a better fit with experiment for _u;ations :K)t longer

than 120 h.

This is the case for the IRAS tests.

7.1 The results are given as computer listing (see example of TML listing

below)............................................. ...

w -R.234420033G • Y 1'Q.0glIR?3"nM53d; (.s.)

_t05 s 10"_0 • 1

............................... o .............................

TIa_(H) M£S* ¢AL('. EIm.J. IMTE

S.O ;I.&611/'L "" 2.34431"1 ** T,1.5?4" -- _?.|400/'1;I.0 t •?341-iE- J. 2. 4900it'lL o. 060%t ,d.|444E't3.0 ;I.?907['| 1.592_'I 7,0991 -- ?.91193P'34.0 2.OdI'_N"L -- ;t.61il08"'i S.O?:lO G.Oe'.W+'II"35.0 a.b651E'1 3.?tlOA"t 5.1761 -- 4.9,'ao_"36,0 2 •O_JS4r"1 2. ?S21UF"1 *- 4.?_33 4. _ | IN_"3?.0 _1.920°-J['1, 2.001 'JE-t 4.07_0 3.6G?_'30.0 2.939SE'1 -- 2. PJG3E't 3.50_ ;J.3,W ._'39.0 ;I. 04(_E'1 2.1_6?1E'1 R.??03 _.91_0E'3

I0,0 _.9bOIE'l R. f1940Lr" _. 10_0 ,", G_."GI "311.0 _.9501Z'1 ;I. 920,TJF"I. |.;1774 ,"..13_G -r'312.0 a,SSOli-I -" ;l. 9437J"1 0.4072 -. :%477E'3!3.0 2.9G741"1 ;I.'_,154F'1 O.Of._ll-- ._.02011"3

14.+ '.9_40"t "" +:009044_.: "" "0.10A'0"014'..'" 1.03'331"01"J040''3L5.O _.0000,'|1,.. 3, 00,3A"1 3. O+;13E'I "0.4337--1. ?30UE',17.11 3.0232'1"1 3.03'9"J1" 1 "0.5:171 -- 1 •1_3011"3liD.I) 3. 032_F01 3.0581E'1 "0.7449 -- I •5_4_1"319,0 3,r,4_,0['1 3.0703E'1 -- "0.935? -- t +UO&E'320,0 3.0_50E'1 *- 3.0040E't "0.94_1 -- 1.4132_"3;ll.O 3.0604E'I 3.0g_E'| -- "l.2.tG3 -- I 35_10[-3;12.0 3 •0f,S";a'" t 3.11,180"1 "t,3707 "- I •,'?.750L"323,0 3. O?')OE'I +- 3 • |34._1r'1 - _, 4?o4ji -- 1,2447E'324,1 3.00031"1 3,13671"t "l.s66s -- 1.19731"325,0 3,1065ur'1 "" 3,14031"1 "1,336, "* 1.1|3_1/'3:'6.1 3. I IIK.31"I 3.151101"; " 1.3WFJ *- I. | 1361"3i?.a 3.134glr'i 3,1?olur'l ' 1,1403 -- t.O?50E'3_10,i) 3.1441L1"1 3. lllL3Jr'| -- "1.1011? -* 1. 041LE'3,19,0 3.155L_ -t 3.19111"I "|. ;10,1,4,-- I. 0i)o41r'331,1 , 3.14411"I " 3.ti)lll'l -- "1.8143 -- 9.??Oalr'4

These listings g_ ve:

a. the power function from which the results %'e_e calculated.

b. the resvlts :

i. in column I: time in hours

2. in column 2: me@sured data (%)

3. in column 3: data calculated from the equation (_)

4. in column 4: difference between data calculated and measured

5. in column 5: the rate in %/h

236

I

1981005626-245

Lm

k__! 7.2 The dlfferent equations calculated for six tests are given in Tables_ 5 and _.

_. We can see from these tables the differences which occur between

: equations corresponding to tests carried out at the same temperature.

'_ These differences and in some cas_s the absence of equation are due

to technical incidents during test_ (i.e.f. ilure of oven, liquid

o nitrogen tank empty, high ncise on the QC24s or failure of LN 2 regula-

tion (LN2 valves still open).these incidents are the cause of variations in the data recorded

(re-evaporation of contaminants from QCMs when the LN_ _s missing for

example) and consequently of me4ification in the calculated equations.These incidents have been particularly f£cquent during the last threetest_.

7.3 The curves measured and calculated (Figure 3) for i_:."and CVCM for

each test indicate a fair correlation for the total mass loss, except

for test 1049 (instabilities due to ice-drift on the vacuum balance),

but are not conclusive in the case of CVCM.

8. PREDICTIONS

From the equations given in Tables 5 and £, we calculated _e % TI.J_,% TML rate, % CVCM and % CVCM rate predictions from 1 to 10-hours.

These predictions are shown in Tables 7, 8 and 9.

9. CONCLUSION

The results obtained show large discrepancies between them. These are

mainly due to technical incidents during the tests and inaccurate

zero during the analysis of recordings.

T,esc differences influenced the equations used to carry out the cal-

culations fur _/]e predictions.

Concerning the form cf equation used to ca:ry out the calculation,

the power function is the best one that could be foun4 at the moment

to fit the experimental data, but it is not in agreement with any

theory that this casts a doubt on the far away extrapolation. To

validate a theory, a lot of long time duration tests (minimum three

hundred hours fo': each sample without incident) should have been run.The predicted outgassing after I0 hours is, however, within a factoraround 2 for TML and I0 for CVCM.

Notwithstanding their lack of accuracy, the results g.ven in this

paper should be sufficient to assess the contamination risk of the

satellite with a reasonable confidence.

237

|

i "" !

1981005626-246

i

c

1042 1043 104_ 1045 10_6 10_7 10_9

time time time time time time time

; Action/failure (h) " (h) (h) (h) (h) (h) (h) I'Mounting samplp -6 -?7 -19 -5 -1 -1 -1

,,i 1Pump dow_ -5 -26 -17 -_ 0 O 0

/_ Jling _,C;is -3 , -24 -15 -2 I C

Hea_ng sample 0 _ 0 0 0 i

! I ", LN2 failcre _?t;.to 105 c _

'{ tankempty i_Oven failure _ 302 and | o,

I 30_ hl

" I ?

390 and ' ,-

_' II _ sgsh : . '_" I I

• Noiseeffects , . _No data on No data on _o _ata on [

_' (unknown origin, I i :QCMs. Too " QCMs. Too QCHs. Too t

probably . _ _i i [ ,'much noise much noise much nois_ [ __ielectrlcal) : _ _ : ._,

valves , ,Valves Valves Valves -ii

Ifroze_ frozen [ frozenfrozen

i, I I ,"I i !

TABLE I.

0 : starting point of data

Test Sample TML CVCM CVCM CVCM I CVCMNumber Number T°C sample -25°C -V5°C -IO0°C -16n°C Observations .]I+ 125°C

X X X t cooling time1042 3 _ 3h t 3h 2 3h

-- duration of

1043 l X t X _ X QCMs before "_c24h 24h sample hearing

1044 2 X X 12

ISh _ 15h long duPa-

-- tic test ilOU_ 4 X X :, (400 h;

2h _2h \_ _I_ --

two testsI0_ _6 X X X X

hav_ been

T sample carrlt,d cut.+ 80°C Ist: C_h ,at

1047 l_S X X X X lO0°C

T sample 2nd: one week; + 100-C at 125°C after

..... recovering at

4. 1049 5 X X X X 65% RH, 25°C

'_ _" samRle duping m ,-6 "

• 125_C than 48 h.-F *

TABLE 2 _,

For the tests 3046 and 104V the cooling of the QCH and heating of the sample have _eea started together

under dry nitrogen at atmo_phezqc pressure before pump do_.,

238

××××-002

}!

' r

P=5xi0-7 P=-_xlC-7 P=1.3xIO -'! P=1.5x10 -7 P=9x19 -8 P=?xl "-8 P=_,xl" "ttoPr toPr tOrP I:01_9 %orP tort t O._'r -_

T2 T3 T4 T5 "J'e T7role (To+l) (To+15) (To+25) (T0 à�l�(TO ˜ä�d�(?oà�l�(To_?7)

Amp _p Amp A_p Amp i. Amp Ao_ !) 1 6 xlO-] 1 _ _ _ I _ _

2 9.6x10 -11 3.2x19 -12 2.9x10 -12 1.7x10 -12 1.9x10 -12 , 1.Ux]n -!2, t _ 12 1.5xlO -11 2.5x10 -13 ....

i 13 .....1_ 6 xlO -11 2.3x10-12 2 x19 -12 1.qx10 -12 1.5x10 -12 t 1.3xl _-12 ,_ x]n "1"_

l 15 1.5x10-11 .... I _16 9 xlO -11 2.7x10 -12 2 xlO -12 1.7x10 -12 1.5xlq -12 I 1.1xl P-12 5 xi _

._-13| ......

17 3 xlO-10 7 xlq -11 6.2x10 -12 q.2x10 -12 3.5x10 -12 2.8xlp -12 1.3x10 -12

_1 _

18 1.3x10-9- 2.2x10 -10 1._xiO -10 1.1xlO -10 _8 xlt_ -11_ 8 xlO -11 a.9xlO "

28 6.9x10 -10 3._x10 -10 2.9x10 -10 2.3x10 -10 2.6x10 -10 1.7x10 -lq 1.6xlq -10

: 32 1.2x10 -10 6 xlO -11 8.9x10 -12 6.6x10 -12 8 xlO -12 _ xlO -12 3.5x10 -12_

_ _ _0 2.1x10 -11 2.5x10 -12 2. lx10 -12 1.9xlO -12 1.TxlO -12 1.1x10 -12 6.8x10 -13

_ 1.5x10 -10 6 xlO -11 _.2x10 -12 3 x10-12 2.2x10 -12 1.3x10 -12 9 xlO -13_ 55 _ x10 -12 - - ' - =

[ 57 q xlO -12 -

57 3 xlO -12 - "-69 3 x10 _ - -

; _ 79 _ x10-12 - " : 'l

TABLE 3

Note: 1o is the time of measurement of the pl.essure after tn_ _n. h=u_ after pump dom, of the sample.

P is the p_ccsure at the time when the recording has\been carried ou_

! Test Sample! restI% 'SamplJ Rate TMLNumber Number Resul_ No. [TML iTemp. !(%/h) _

i .lq_.2 3 HC : 5._ x 10-5 - 10_2| 0.036601125°C [ 1._567x10-2T -0"963_0,0.2980(T)

I E : 3.1 x 10-5 , ,8.5 x l_--_JS--_ 1003 _ 0.029761 125°C ! 1.01_5xln-2T -0"9702_ :

i 10,3 1 HC : 2.5 x 10-5 I ! ,_io.,og,T,E : 1.5 x 10 -5

•_ _ x 10-_ g 10_qi 0.02662 125°C 7._772x1C'3T-0"9733-8 '_.- _ :0.280(T)- _ ]_q 2 HC : 8._ _ 10-5

_ E : 2.6 x 10-5 1005 0.5523 1250C 1.0529X10"2T "0"90_77L L 1.11 x 10-_" g ,0.3355(T) ; _

_ _ 10_|5 _ HC : 1.25 x i0 -_ 10_6 0.09162 80°C 2.1475xlO'2T "0"90838 ;_ E : 0.70 x i0"_ ;-

_ 1.95 x 10"W g LO_7I 0.097_2 IooOc 1.76_8xlO'2T "0'90260

V 10_6 6 HC : 2.0q x 10-5 I0.1812(T) ;

_ E : 8.90 x 10-6 L0_9; 0.02330!125°C __.375 xlO'2T "0"97670,.:- 2._3 X I0"_ g :

i 0'1873(T) ilOa7 5 HC:2.89x lO"5 1 __" E : 1.13 x 10 -5 TABLE 5 "

_. _. 02 x 10-r_ g _ ,

HC : hydrocarbon TABLE q 1" :. - _steP

"_. 7 _::e results have been obtained according todocument PSS-15/QRM-OST,

, 239 i

[,

XXXX-O03

Test % CVCM TemP. I Rate CVCNNumber °/C (t/h)

.: 0.1916 -0.808_1042 ....11.7T -25n nn n 2.1210X10-"T

' 10_2 0.2311 -750.0012,6T 2.8795xlO-gT -0.7689

0.3256 /i0_3 O.00C_O_T -i00 2.9317x10-qT -n.67u_

0.3570 l_ 10_3 0.001012T -160 4.0286x10-"T -0.6,30

0.2987: _ lO_q -100 -0.7013

0.001161T .3._679x10-_T_ 0.3276

I0_ 0.001232T -160 _.0360x10-_T -0.672_ •

_ 0.3513_Y 10,5 0.000611T -25 2.1,6_x10-_T -0.6_87

0.3026 -75 -0.697_ !_] i0_5 0.000899T 2.7203x10-_T :

_ lO.6 ND -TS .D )=

10_6 ND -75 ND = t :

i 1047 ND -25 ND

i 1047 ND -25 ND?

!

IO_9 ND ND

i0_9 ND ND

TABLE 6

=

! Time test i0a2 test lOq3 test IOq4 test 1045 test ] test I_47 test 10v9

1 2.gSx10 -1 3.USx10 -1 2.80x10 "1 2.96x10 "1 2.3_x_ 1.81x10 -1 1.88x10 "1

i0 3.2_xi0 -1 3.65xi0 -I 2.98xi0 -I 3.81xi0 -I 2.89xi0 2.26x1_. "l 1.98x19 -I

102 3.52x10 -1 3.91x10 "1 3.17x10 -1 u,. 32x10 -1 3.57x10 "1 2.83x10 -1 2.09x10 -1

1,, 3: 3.83xi0 -I q. 18xlO -I 3.37xi0 -I 4.91xI0 -I _.13xlO -I 3.55x10 "I 2._IxlO -I

; 10_ u. 17x10"1 _._8x10 -1 3,58x10 "1 5.58x10 -1 5._5x10 -1 _,._qxlO -1 2.33xi0-I

I _5 _.5gx ._-1 _.SOxlO "I 3.81xi0 "I 6.33xl _'I 6.73xi0 "I 5.56xi0 -I 2.U6xlO "I

_ TABLE 7

Time i test I0_2 test ,C_3 test lO_q test I0_45 test IOq5 test IOw7 test 10_9

;_ I l I._56"×I(?"2 r.ol_x19 _ 7.q72xlo"3 1.853xi0 "2 2.147xi0 "2 1.76_x10-2 _.375xI0 "3

_ lt_ i 1.585 x10"3 !.086xI0"3 7.9hqxlO-q 2.103xi0 -3 2.651xi0 -3 2.208xi0 -3 q.£16xlO -q

102 i 1.72q xlo"g 1.163xI0"_ 8.44 xlO"5 2.38 xlO "_ 3.273xI0 "_ _2.76_x10"_ g.871xlC -5

.i i, I 1.875 xl0-5 1.296x10"5 8.981xi0"6 2.712xi0"5 _.0,2xi0 "5 3.q59x10 "5 5.1kOxlq"6

It" I 0_1 xlJ -6 1,33_x1c -6 9,53 xlo -7 3,08_x10 -6 4,992x10 -6 _.329x10 -6 9,lq.xlo -7=

2.22 xl0 -7 1,g3 xl0 -7 1.02 xl0 "8 /_.50 xl0" 7 6.16 xl0 "7 5.q2 xl0 "7 5.7 xl0 "8

TABLE 8 •

• Pooa

!l ! --24 0 "--

!

××××-004

t

t

F •

i! Time test 1042 test 1042 test 1042 test log2 test 110_3 test I0_3' test log3! "test lOq3

|: (hrs) at -25 C at -25°C at -75°C at -75°C at -lO0°C at -lO0°C at -160°C at -160°C

. , % CVCM %/h % CVCM %lh %.CVCM %lh % CVCM %lh

i l,llxlO "3 2.12xI0 -4 1.25xi0 -3 2.87xI0 "_ 9.00xlO -_ 2.93xI0 -_ 1.01x10"3 _.02xlO -4

i0 1.72xi0 "3 3.29xi0 -5 2.12xi0 "3 _.9 xlO "5 1,9 xlO -3 6.2 xlO -5 2.30x10 "3 9.16XI0 "5

102 2.67xi0 -3 5.12xi0 -6 3.61xi0 "3 B.3_xlO -6 _,03x10 -3 1.3 xlO "5 5.23xi0 "3 2.08XI0 -5103 _.16xlO "3 9.gbxlO "7 6.14xi0 -3 1.42x10 -6 8.53xi0 "3 2.77xi0 "6 1.19x10 °2 _.7_xlO -6

i 104 6._6x10 -3 1.23xI0 "_ 1.05x10 -2 2.41x10 -7 l.BOxlO -2 5.I$xi0 "7 2.71xi0 -_ 1.08xi0 "6

105 i xlO "2 1.92xi0 -8 1.78xi0 -2 4.11x10 -8 3.8_x10 -2 i.?4x10 "7 6.16xi0 "2 2.qSxlO -7

Time test I0_ test I0_ test i0_ test 1044 test i0_5 test 1045 test I0_5 test 10_b(hrs at -IOOOC at -lO0°C at -160°C at -160°C at -25°C at -25°C at -75°C at -75°C

% CVCM %/h % CVCM %/h % CVCM %/h % CVCM %/h

i "i 1.16xlO -3 3.h6xlO -_ 1.23xi0 -3 u,O3xlO -4 6.11xlO -g 2.15xlO "g g xlO "_ 2,72xI0 -4 :

h) 2.31xi0 "3 6.9 xlO -5 2.62xi0 -3 B,5Bxl_ "5 1 37xi0 -3 g.81x10 " l.d xlO -3 l,l xlO "_ •102 _.6 xlO "3 1.37xi0 -5 5.57xI0 -_ 1,82xi0 -5 3.08xln "3 l.O8xlO -3 3.62×I0 -3 I.i ×I0 -5

103 q.13xlO "3 2.73xi0 -6 I.IBxlO " 3.82xi0 -e 6.q xlO "5 ?.h3xlO -6 7.27x10 -3 2.2 xlO -b ""

-2 -'2 -7 -_ -7 -2 -7I0 _ 1.91X]O 5._3x10 -7 2.51x10 8.2_XI0 l.bSxlO _ _._Sxl' l._oxlO _.t_IxlO

i n5 3.61xI0 -_ 1 08xlO -7 5 3_xlO -_ 1.75xi0 -} 3.gSxlO -_ 1.22x10 "7 2.93xI, " 8.Rbxlt -8

&

TABLE 9

Those _esu!ts a_e plotted on the following curves.

If

. !•

• I 241

5 _\ •

××××-005

TML prediction rates

aUn,e a,on,_t,_. N t#an (q_.,_.l}

_ lP'ONq (g'fe_e(I#i

ii,".,.'_,_ CVC._Ipredlcglon242rates

XXXX-O06